Applying machine learning to semiconductor manufacturing

Irani, Keki B.; Cheng, Jie; Fayyad, Usama M.; Qian, Zhaogang

doi:10.1109/64.193054

Cited by 64 publications

(23 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Malkoff (1987) used temporal data in performing fault diagnosis in a subsystem of a Navy Ship propulsion system. In the 1990s, Irani et al (1993) generalised an ID3 algorithm that predicts the outcome of future experiments under various, more general conditions. Their systems have been used successfully in various semiconductor manufacturing applications in both diagnosis and process modelling.…”

Section: Literature Reviewmentioning

confidence: 99%

Enhancing product quality of a process

Çiflikli

Kahya-Özyirmidokuz

2012

Industrial Management &Amp; Data Systems

View full text Add to dashboard Cite

If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information. About Emerald www.emeraldinsight.comEmerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services.Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation. AbstractPurpose -Data mining (DM) is used to improve the performance of manufacturing quality control activity, and reduces productivity loss. The purpose of this paper is to discover useful hidden patterns from fabric data to reduce the amount of defective goods and increase overall quality. Design/methodology/approach -This research examines the improvement of manufacturing process via DM techniques. The paper explores the use of different preprocessing and DM techniques (rough sets theory, attribute relevance analysis, anomaly detection analysis, decision trees and rule induction) in carpet manufacturing as the real world application problem. SPSS Clementine Programme, Rosetta Toolkit, ASP (Active Server Pages) and VBScript programming language are used. Findings -The most important variables of attributes that are effective in product quality are determined. A decision tree (DT) and decision rules are generated. Therefore, the faults in the process are detected. An on-line programme is generated and the model's results are used to ensure the prevention of faulty products. Research limitations/implications -In time, this model will lose its validity. Therefore, it must be redeveloped periodically. Practical implications -This study's productivity can be increased especially with the help of artificial intelligence technology. This research can also be applied to different industries. Originality/value -The size and complexity of data make extraction difficult. Attribute relevance analysis is proposed for the selection of the attribute variables. The knowledge discovery in databases process is used. In addition, the system can be followed on-line with this interactive ability.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Enhancing product quality of a process

Çiflikli

Kahya-Özyirmidokuz

2012

Industrial Management &Amp; Data Systems

View full text Add to dashboard Cite

show abstract

“…In its purest and most ambitious form, our objective is to predict the final outcome of each wafer in terms of critical functional characteristics. Months may pass before a chip is completed, hence the great interest in mining data prior to final testing [9], [1], [5]. Moreover, if such an endeavor were to be successful, it would greatly enhance manufacturing productivity.…”

Section: Introductionmentioning

confidence: 98%

Improving quality control by early prediction of manufacturing outcomes

Weiss

Dhurandhar

Baseman

2013

Proceedings of the 19th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining

View full text Add to dashboard Cite

We describe methods for continual prediction of manufactured product quality prior to final testing. In our most expansive modeling approach, an estimated final characteristic of a product is updated after each manufacturing operation. Our initial application is for the manufacture of microprocessors, and we predict final microprocessor speed. Using these predictions, early corrective manufacturing actions may be taken to increase the speed of expected slow wafers (a collection of microprocessors) or reduce the speed of fast wafers. Such predictions may also be used to initiate corrective supply chain management actions. Developing statistical learning models for this task has many complicating factors: (a) a temporally unstable population (b) missing data that is a result of sparsely sampled measurements and (c) relatively few available measurements prior to corrective action opportunities. In a real manufacturing pilot application, our automated models selected 125 fast wafers in real-time. As predicted, those wafers were significantly faster than average. During manufacture, downstream corrective processing restored 25 nominally unacceptable wafers to normal operation.

show abstract

“…However, many opportunities for yield improvement are far more subtle [5,6]. Months may pass before a chip is completed, hence the great interest in mining data prior to final testing [7][8][9]. Some chips will fail these tests, and the percentage of successfully manufactured chips, especially during the early stages of manufacturing, may be significantly below 100%.…”

Section: Introductionmentioning

confidence: 99%

Rule-based data mining for yield improvement in semiconductor manufacturing

et al. 2009

View full text Add to dashboard Cite

We describe an automated system for improving yield, power consumption and speed characteristics in the manufacture of semiconductors. Data are continually collected in the form of a history of tool usage, electrical and other real-valued measurements-a dimension of tens of thousands of features. Unique to this approach is the inference of patterns in the form of binary regression rules that demonstrate a significantly higher or lower performance value for tools relative to the overall mean for that manufacturing step. Results are filtered by knowledge-based constraints, increasing the likelihood that empirically validated rules will prove interesting and worth further investigation. This system is currently installed in the IBM 300 mm fab, manufacturing game chips and microprocessors. It has detected numerous opportunities for yield and performance improvement, saving many millions of dollars.

show abstract

Applying machine learning to semiconductor manufacturing

Cited by 64 publications

References 4 publications

Enhancing product quality of a process

Enhancing product quality of a process

Improving quality control by early prediction of manufacturing outcomes

Rule-based data mining for yield improvement in semiconductor manufacturing

Contact Info

Product

Resources

About